Con­fir­ma­tion bias: how errors in reasoning can be used in marketing

Women don’t know how to park a car, left-handed people are more creative than right-handed people, more and more people meditate – however absurd or obvious an as­sump­tion may seem to you, everyone has certain strong beliefs they are not willing to let go of without a fight. You’ve probably had at least a few con­ver­sa­tions where the person you were talking to stub­born­ly main­tained their opinion, even though you con­front­ed them with facts that exposed their point of view as ob­jec­tive­ly false. People are inclined to downplay or ignore such in­for­ma­tion entirely, while in everyday life they always latch onto what appears like evidence for the validity of their arguments.

These types of sit­u­a­tions are examples of con­fir­ma­tion bias at work. This psy­cho­log­i­cal flaw in reasoning distorts our per­cep­tion. It ensures that we only se­lec­tive­ly as­sim­i­late new in­for­ma­tion and adhere to con­vic­tions that we already held to begin with.

Human thinking is anything but objective. Rather, it is subject to diverse cognitive dis­tor­tions. That’s why when pro­cess­ing in­for­ma­tion, we regularly and sys­tem­at­i­cal­ly make mistakes.

The psy­chol­o­gist Peter Wason demon­strat­ed the existence of the reasoning error in the 1960s as the first of several ex­per­i­ments.

In one of these ex­per­i­ments (2-4-6 task) the test subjects were asked to identify according to which rule number sequences of three numbers had been arranged. The number sequence 2-4-6 was given to par­tic­i­pants, after which they could specify their own number com­bi­na­tions that they assumed conformed with the rule. Lastly, they were asked to state the rule that they believed to have iden­ti­fied. The results, according to Wason, showed that the test subjects were inclined to only test those number sequences that supported their own as­sump­tion.

Wason was also able to observe this fallacy in other ex­per­i­ments – e.g. during a selection test/Four-Card-Problem. In the following decades, further research regarding this subject resulted in numerous new findings and the term “con­fir­ma­tion bias” is now used to denote an entire series of thinking and memory patterns. Below are some examples:

• In­for­ma­tion ac­qui­si­tion: People only gather in­for­ma­tion that supports their own as­sump­tions (Wason’s de­f­i­n­i­tion of con­fir­ma­tion bias).
• Memory: People only recall in­for­ma­tion that is com­pat­i­ble with their own opinion.
• Framing: People interpret new in­for­ma­tion in a way that is in line with the attitude they have held up to now.
• Reassess: People tend to pass on op­por­tu­ni­ties to question their own opinion.
• Reject: People reject in­for­ma­tion that is not in agreement with their own con­vic­tions.

Research on con­fir­ma­tion bias and whether we can free ourselves from it have occupied psy­chol­o­gists for decades. And for good reason. This is because the cognitive dis­tor­tion can have serious and sometimes dangerous con­se­quences.

If a doctor doesn’t examine a patient thor­ough­ly, for example, because they are known to be a hypochon­dri­ac and their symptoms are not to be taken seriously, the doctor could fail to identify a serious illness in time, ul­ti­mate­ly leading to the patient’s death.

But it doesn’t have to be this dramatic. Even when watching or reading the news, we disregard in­for­ma­tion that is not in line with our beliefs due to con­fir­ma­tion bias. In addition to this, al­go­rithms in social media only present us with facts that are in agreement with our worldview. This widens the gap between different political camps, making it harder and harder to rectify rad­i­cal­ized opinions.

The existence of this reasoning error is widely regarded as certain among sci­en­tists. The extent to which people are actually in­flu­enced in their decisions by con­fir­ma­tion bias, however, to this day remains disputed. Gary Klein, for example, sees central weak­ness­es in the research.

The fact that people are inclined to adhere to their con­vic­tions and un­con­scious­ly confirm their existing as­sump­tions has been used suc­cess­ful­ly by companies in their marketing for many years. And you can easily follow their lead.

In essence, it is a question of finding out which implicit as­sump­tions your customers have and con­firm­ing these via your marketing com­mu­ni­ca­tions. In doing so you gain their approval, earn their trust and are one step closer to com­plet­ing the sale.

German car man­u­fac­tur­ers deliver quality, French wines are first rate, and Silicon Valley start-ups develop brilliant tech­no­log­i­cal in­no­va­tions. Many people hold beliefs related to in­dus­tries or in­di­vid­ual brands.

If there are positive as­sump­tions about your company, or your industry has a rep­u­ta­tion of trust­wor­thi­ness, you can seize upon these as­sump­tions in your marketing com­mu­ni­ca­tions. People will readily believe these com­mu­ni­ca­tions as they will fit like puzzle pieces in your company or industry’s public image. Marketing is often more efficient when it builds upon positive pre­sump­tions.

What if your company is still fairly unknown or people do not have positive as­so­ci­a­tions about your industry in mind? This means you have a specific issue for which people hope to find a solution from you – and this lets you play off of con­fir­ma­tion bias. People want to believe that there is a solution for their problem. They are drawn to arguments that promise them a feasible remedy for it.

The more familiar you are with your customers’ pain points, the more clearly you will be able to address these in your sales ar­gu­men­ta­tion, and the more likely your customers are to select your company over a com­peti­tor. However, con­fir­ma­tion bias has another impact: customers are often not fully aware of their problem. By ad­dress­ing their paint points, you can confirm un­con­scious fears, only to then present your suitable solution to a customer who is now receptive.

On the Internet, customers often find it difficult to judge whether a seller is trust­wor­thy. And without this trust, there can be no sale. Even when customers are in­ter­est­ed in a product, they are con­scious­ly or un­con­scious­ly looking for in­di­ca­tors to confirm they have found a serious and competent seller.

Satisfy your customer’s un­con­scious wish and publish positive customer tes­ti­mo­ni­als or success stories to prove that your product is worth their money and that it is sensible to do business with you. By doing this, you give in­ter­est­ed parties the necessary nudge to make the final click in their shopping cart.

The larger the purchase, the stronger the customer’s desire to ensure that their money is being well invested. To address this, companies have many options. Why not send a thank you email after the trans­ac­tion in which you suggest a few practical tips for using your product, offer a discount for another product, or attach a free goodie?

And even if the purchase occurred less recently, it is worth­while investing in your re­la­tion­ship with your customers – es­pe­cial­ly when using sub­scrip­tion models. Make your customers feel good about using your services wherever you can. Mailchimp provides a good example of this: As soon as one of their users sends out an email campaign, they are con­grat­u­lat­ed with a quirky virtual high five. This lets them leave the ap­pli­ca­tion with a smile on their face and feeling affirmed in their choice of an email provider.

The tips we that we have provided up to this point address how you can implement con­fir­ma­tion bias in your marketing so that your com­mu­ni­ca­tions with the customer are more ef­fec­tive­ly planned. Many marketing de­part­ments, however, forget that they them­selves also are liable to con­fir­ma­tion bias.

Reflect regularly on which implicit as­sump­tions underlie your marketing campaigns. The risk is large that people utilize their marketing tools in such a way that the analysis confirms their own beliefs. Pointedly examine the counter-thesis and try al­ter­na­tive ap­proach­es, for example, through A/B tests.

Con­fir­ma­tion bias is one of the cognitive dis­tor­tions that is the easiest to com­pre­hend. Everyone is familiar with the filter bubbles in social media and knows how difficult it is to dismiss some of their con­vic­tions. Using these facts to your advantage in marketing, unlike the IKEA effect or the endowment effect, is for many, however, less obvious. At the same time, both large and small brands have for a long time con­scious­ly toyed with con­fir­ma­tion bias.

If you want to optimize your marketing, you should take common reasoning errors into account for your marketing measures: for example, the anchoring effect, selection bias and the decoy effect. The larger your arsenal of psy­cho­log­i­cal mech­a­nisms, the more ef­fi­cient­ly you can address potential customers in their journey toward becoming actual customers.